Geoff Hugo
@gdhugo.bsky.social
medical physicist, cardiac radioablation, medical image analysis, AI/ML quality assurance
Reposted by Geoff Hugo
Love this. Bold face numbers in a table don't cut it:
"For >60% of papers, the 2nd-ranked method was within the CI of the 1st-ranked method. Current publications typically don't provide sufficient evidence to support which models could be translated into clinical practice" arxiv.org/abs/2409.17763
"For >60% of papers, the 2nd-ranked method was within the CI of the 1st-ranked method. Current publications typically don't provide sufficient evidence to support which models could be translated into clinical practice" arxiv.org/abs/2409.17763
Confidence intervals uncovered: Are we ready for real-world medical imaging AI?
Medical imaging is spearheading the AI transformation of healthcare. Performance reporting is key to determine which methods should be translated into clinical practice. Frequently, broad conclusions ...
arxiv.org
January 30, 2025 at 2:14 AM
Love this. Bold face numbers in a table don't cut it:
"For >60% of papers, the 2nd-ranked method was within the CI of the 1st-ranked method. Current publications typically don't provide sufficient evidence to support which models could be translated into clinical practice" arxiv.org/abs/2409.17763
"For >60% of papers, the 2nd-ranked method was within the CI of the 1st-ranked method. Current publications typically don't provide sufficient evidence to support which models could be translated into clinical practice" arxiv.org/abs/2409.17763
Reposted by Geoff Hugo
Not every study needs to have a power analysis, but every study needs a sample size justification. I discuss 6 approaches, and 6 ways to think about which effect sizes are of interest in the study you are planning.
online.ucpress.edu/collabra/art...
online.ucpress.edu/collabra/art...
December 18, 2024 at 5:15 PM
Not every study needs to have a power analysis, but every study needs a sample size justification. I discuss 6 approaches, and 6 ways to think about which effect sizes are of interest in the study you are planning.
online.ucpress.edu/collabra/art...
online.ucpress.edu/collabra/art...
Reposted by Geoff Hugo
hey bluesky, in case anyone's interested, I wrote this for my students about why I don't want them to use anything AI in my classes. It interrupts the course calendar part of the syllabus, coming immediately after the first day there's a writing assignment assigned.
docs.google.com/document/d/1...
docs.google.com/document/d/1...
A note on AI usage
A note on AI usage I genuinely want you to succeed in this class. One of my very favorite things about my job as a professor is when students are succeeding and I feel like I helped. I want you to suc...
docs.google.com
December 12, 2024 at 2:25 AM
hey bluesky, in case anyone's interested, I wrote this for my students about why I don't want them to use anything AI in my classes. It interrupts the course calendar part of the syllabus, coming immediately after the first day there's a writing assignment assigned.
docs.google.com/document/d/1...
docs.google.com/document/d/1...
Reposted by Geoff Hugo
Reposted by Geoff Hugo
